̽��ֱ�� of Cambridge - computing

Act now to prevent uncontrolled rise in carbon footprint of computational science

cjb250 — Mon, 26 Jun 2023 09:04:03 +0000

Writing in Nature Computational Science, researchers from the Department of Public Health and Primary Care at the ̽��ֱ�� of Cambridge argue that the scientific community needs to act now if it is to prevent a potentially uncontrolled rise in the carbon footprint of computational science as data science and algorithms increase in usage.

Dr Loïc Lannelongue, who is a research associate in biomedical data science and a postdoctoral associate at Jesus College, Cambridge, said: “Science has transformed our understanding of the world around us and has led to great benefits to society. But this has come with a not-insignificant – and not always well understood – impact on the environment. As scientists – as with people working in every sector – it’s important that we do what we can to reduce the carbon footprint of our work to ensure that the benefits of our discoveries are not outweighed by their environmental costs.”

Recent studies have begun to explore the environmental impacts of scientific research, with an initial focus on scientific conferences and experimental laboratories. For example, the 2019 Fall Meeting of the American Geophysical Union was estimated to emit 80,000 tons of CO2e* (tCO2e), equivalent to the average weekly emissions of the city of Edinburgh, UK. ̽��ֱ��annual carbon footprint of a typical life science laboratory has been estimated to be around 20 tCO2e.

But there is one aspect of research that often gets overlooked – and which can have a substantial environmental impact: high performance and cloud computing.

In 2020, the Information and Communication Technologies sector was estimated to have made up between 1.8% and 2.8% of global greenhouse gas emissions – more than aviation (1.9%). In addition to the environmental effects of electricity usage, manufacturing and disposal of hardware, there are also concerns around data centres’ water usage and land footprint.

Professor Michael Inouye said: “While the environmental impact of experimental ‘wet’ labs is more immediately obvious, the impact of algorithms is less clear and often underestimated. While new hardware, lower-energy data centres and more efficient high performance computing systems can help reduce their impact, the increasing ubiquity of artificial intelligence and data science more generally means their carbon footprint could grow exponentially in coming years if we don’t act now.”

To help address this issue, the team has developed GREENER (Governance, Responsibility, Estimation, Energy and embodied impacts, New collaborations, Education and Research), a set of principles to allow the computational science community to lead the way in sustainable research practices, maximising computational science’s benefit to both humanity and the environment.

Governance and Responsibility

Everyone involved in computational science has a role to play in making the field more sustainable: individual and institutional responsibility is a necessary step to ensure transparency and reduction of greenhouse gas emission.

For example, institutions themselves can be key to managing and expanding centralised data infrastructures, and in ensuring that procurement decisions take into account both the manufacturing and operational footprint of hardware purchases. IT teams in high performance computing (HPC) centres can play a key role, both in terms of training and helping scientists monitor the carbon footprint of their work. Principal Investigators can encourage their teams to think about this issue and give access to suitable training. Funding bodies can influence researchers by requiring estimates of carbon footprints to be included in funding applications.

Estimate and report the energy consumption of algorithms

Estimating and monitoring the carbon footprint of computations identifies inefficiencies and opportunities for improvement.

User-level metrics are crucial to understanding environmental impacts and promoting personal responsibility. ̽��ֱ��financial cost of running computations is often negligible, particularly in academia, and scientists may have the impression of unlimited and inconsequential computing capacity. Quantifying the carbon footprint of individual projects helps raise awareness of the true costs of research.

Tackling Energy and embodied impacts through New collaborations

Minimising carbon intensity – that is, the carbon footprint of producing electricity – is one of the most immediately impactful ways to reduce greenhouse gas emissions. This could involve relocating computations to low-carbon settings and countries, but this needs to be done with equity in mind. Carbon intensities can differ by as much as three orders of magnitude between the top and bottom performing high-income countries (from 0.10 gCO2e/kWh in Iceland to 770 gCO2e/kWh in Australia).

̽��ֱ��footprint of user devices is also a factor: one estimate found that almost three-quarters (72%) of the energy footprint of streaming a video to a laptop is from the laptop, with 23% used in transmission and a mere 5% at the data centre.

Another key consideration is data storage. ̽��ֱ��carbon footprint of storing data depends on numerous factors, but the life cycle footprint of storing one terabyte of data for a year is of the order of 10 kg CO2e. This issue is exacerbated by the duplication of such datasets in order for each institution, and sometimes each research group, to have a copy. Large (hyperscale) data centres are expected to be more energy efficient, but they may also encourage unnecessary increases in the scale of computing (the ‘rebound effect’).

Education and Research

Education is essential to raise awareness of the issues with different stakeholders. Integrating sustainability into computational training courses is a tangible first step toward reducing carbon footprints. Investing in research that will catalyse innovation in the field of environmentally sustainable computational science is a crucial role for funders and institutions to play.

Recent studies found that the most widely-used programming languages in research, such as R and Python, tend to be the least energy efficient ones, highlighting the importance of having trained Research Software Engineers within research groups to ensure that the algorithms used are efficiently implemented. There is also scope to use current tools more efficiently by better understanding and monitoring how coding choices impact carbon footprints.

Dr Lannelongue said: “Computational scientists have a real opportunity to lead the way in sustainability, but this is going to involve a change in our culture and the ways we work. There will need to more transparency, more awareness, better training and resources, and improved policies.

“Cooperation, open science, and equitable access to low-carbon computing facilities will also be crucial. We need to make sure that sustainable solutions work for everyone, as they frequently have the least benefit for populations, often in low- and middle-income countries, who suffer the most from climate change.”

Professor Inouye added: “Everyone in the field – from funders to journals to institutions down to individuals – plays an important role and can, themselves, make a positive impact. We have an immense opportunity to make a change, but the clock is ticking.”

̽��ֱ��research was a collaboration with major stakeholders including Health Data Research UK, EMBL-EBI, Wellcome and UK Research and Innovation (UKRI).

*CO2e, or CO2-equivalent, summarises the global warming impacts of a range of greenhouse gases and is the standard metric for carbon footprints, although its accuracy is sometimes debated.

Reference
Lannelongue, L et al. GREENER principles for environmentally sustainable computational science. Nat Comp Sci; 26 June; DOI: 10.1038/s43588-023-00461-y

For more information on energy-related research in Cambridge, please visit Energy IRC, which brings together Cambridge’s research knowledge and expertise, in collaboration with global partners, to create solutions for a sustainable and resilient energy landscape for generations to come.

Cambridge scientists have set out principles for how computational science – which powers discoveries from unveiling the mysteries of the universe to developing treatments to fight cancer to improving our understanding of the human genome, but can have a substantial carbon footprint – can be made more environmentally sustainable.

Science has transformed our understanding of the world around us and has led to great benefits to society. But this has come with a not-insignificant – and not always well understood – impact on the environment

Loic Lannelongue

geralt

Binary world

̽��ֱ��text in this work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License. Images, including our videos, are Copyright © ̽��ֱ�� of Cambridge and licensors/contributors as identified. All rights reserved. We make our image and video content available in a number of ways – as here, on our main website under its Terms and conditions, and on a range of channels including social media that permit your use and sharing of our content under their respective Terms.

Yes

Licence type:

Public Domain

Mathematical paradox demonstrates the limits of AI

sc604 — Thu, 17 Mar 2022 16:05:06 +0000

Like some people, AI systems often have a degree of confidence that far exceeds their actual abilities. And like an overconfident person, many AI systems don’t know when they’re making mistakes. Sometimes it’s even more difficult for an AI system to realise when it’s making a mistake than to produce a correct result.

Researchers from the ̽��ֱ�� of Cambridge and the ̽��ֱ�� of Oslo say that instability is the Achilles’ heel of modern AI and that a mathematical paradox shows AI’s limitations. Neural networks, the state-of-the-art tool in AI, roughly mimic the links between neurons in the brain. ̽��ֱ��researchers show that there are problems where stable and accurate neural networks exist, yet no algorithm can produce such a network. Only in specific cases can algorithms compute stable and accurate neural networks.

̽��ֱ��researchers propose a classification theory describing when neural networks can be trained to provide a trustworthy AI system under certain specific conditions. Their results are reported in the Proceedings of the National Academy of Sciences.

Deep learning, the leading AI technology for pattern recognition, has been the subject of numerous breathless headlines. Examples include diagnosing disease more accurately than physicians or preventing road accidents through autonomous driving. However, many deep learning systems are untrustworthy and easy to fool.

“Many AI systems are unstable, and it’s becoming a major liability, especially as they are increasingly used in high-risk areas such as disease diagnosis or autonomous vehicles,” said co-author Professor Anders Hansen from Cambridge’s Department of Applied Mathematics and Theoretical Physics. “If AI systems are used in areas where they can do real harm if they go wrong, trust in those systems has got to be the top priority.”

̽��ֱ��paradox identified by the researchers traces back to two 20th century mathematical giants: Alan Turing and Kurt Gödel. At the beginning of the 20th century, mathematicians attempted to justify mathematics as the ultimate consistent language of science. However, Turing and Gödel showed a paradox at the heart of mathematics: it is impossible to prove whether certain mathematical statements are true or false, and some computational problems cannot be tackled with algorithms. And, whenever a mathematical system is rich enough to describe the arithmetic we learn at school, it cannot prove its own consistency.

Decades later, the mathematician Steve Smale proposed a list of 18 unsolved mathematical problems for the 21st century. ̽��ֱ��18th problem concerned the limits of intelligence for both humans and machines.

“ ̽��ֱ��paradox first identified by Turing and Gödel has now been brought forward into the world of AI by Smale and others,” said co-author Dr Matthew Colbrook from the Department of Applied Mathematics and Theoretical Physics. “There are fundamental limits inherent in mathematics and, similarly, AI algorithms can’t exist for certain problems.”

̽��ֱ��researchers say that, because of this paradox, there are cases where good neural networks can exist, yet an inherently trustworthy one cannot be built. “No matter how accurate your data is, you can never get the perfect information to build the required neural network,” said co-author Dr Vegard Antun from the ̽��ֱ�� of Oslo.

̽��ֱ��impossibility of computing the good existing neural network is also true regardless of the amount of training data. No matter how much data an algorithm can access, it will not produce the desired network. “This is similar to Turing’s argument: there are computational problems that cannot be solved regardless of computing power and runtime,” said Hansen.

̽��ֱ��researchers say that not all AI is inherently flawed, but it’s only reliable in specific areas, using specific methods. “ ̽��ֱ��issue is with areas where you need a guarantee, because many AI systems are a black box,” said Colbrook. “It’s completely fine in some situations for an AI to make mistakes, but it needs to be honest about it. And that’s not what we’re seeing for many systems – there’s no way of knowing when they’re more confident or less confident about a decision.”

“Currently, AI systems can sometimes have a touch of guesswork to them,” said Hansen.“You try something, and if it doesn’t work, you add more stuff, hoping it works. At some point, you’ll get tired of not getting what you want, and you’ll try a different method. It’s important to understand the limitations of different approaches. We are at the stage where the practical successes of AI are far ahead of theory and understanding. A program on understanding the foundations of AI computing is needed to bridge this gap.”

“When 20th-century mathematicians identified different paradoxes, they didn’t stop studying mathematics. They just had to find new paths, because they understood the limitations,” said Colbrook. “For AI, it may be a case of changing paths or developing new ones to build systems that can solve problems in a trustworthy and transparent way, while understanding their limitations.”

̽��ֱ��next stage for the researchers is to combine approximation theory, numerical analysis and foundations of computations to determine which neural networks can be computed by algorithms, and which can be made stable and trustworthy. Just as the paradoxes on the limitations of mathematics and computers identified by Gödel and Turing led to rich foundation theories — describing both the limitations and the possibilities of mathematics and computations — perhaps a similar foundations theory may blossom in AI.

Matthew Colbrook is a Junior Research Fellow at Trinity College, Cambridge. Anders Hansen is a Fellow at Peterhouse, Cambridge. ̽��ֱ��research was supported in part by the Royal Society.

Reference:
Matthew J Colbrook, Vegard Antun, and Anders C Hansen. ‘ ̽��ֱ��difficulty of computing stable and accurate neural networks – On the barriers of deep learning and Smale’s 18th problem.’ Proceedings of the National Academy of Sciences (2022). DOI: 10.1073/pnas.2107151119

Humans are usually pretty good at recognising when they get things wrong, but artificial intelligence systems are not. According to a new study, AI generally suffers from inherent limitations due to a century-old mathematical paradox.

There are fundamental limits inherent in mathematics and, similarly, AI algorithms can’t exist for certain problems

Matthew Colbrook

Yuichiro Chino

Binary data wave

̽��ֱ��text in this work is licensed under a Creative Commons Attribution 4.0 International License. Images, including our videos, are Copyright © ̽��ֱ�� of Cambridge and licensors/contributors as identified. All rights reserved. We make our image and video content available in a number of ways – as here, on our main website under its Terms and conditions, and on a range of channels including social media that permit your use and sharing of our content under their respective Terms.

Yes

Cambridge-built carbon credit marketplace will support reforestation

sc604 — Wed, 10 Nov 2021 11:58:35 +0000

̽��ֱ��Cambridge Centre for Carbon Credits (4C) - based in the Department of Computer Science and Technology, and the ̽��ֱ�� of Cambridge Conservation Research Institute - has two primary goals: to support students and researchers in the relevant areas of computer science, environmental science, and economics; and to create a decentralised marketplace where purchasers of carbon credits can confidently and directly fund trusted nature-based projects.

̽��ֱ��Centre will build its decentralised marketplace on the energy-efficient Tezos blockchain because it operates sustainably and allows third parties to verify all transactions, in line with the Centre’s vision to support a sustainable future through technology. ̽��ֱ��goal of the marketplace is to exponentially increase the number of real nature-based conservation and restoration projects by channelling funding towards them via market-based instruments.

Nature-based solutions, particularly forests, have a vital role to play in mitigating the worst effects of climate change. Pressure is mounting from governments and the public to rapidly roll out a global programme of well-executed nature-based solutions (NbS) to sequester several gigatons of carbon each year and protect biodiversity. However, current NbS projects are hampered by chronic underfunding.

“Current accreditation systems that measure and report the value of carbon and related benefits like biodiversity conservation and poverty reduction rendered by NbS are costly, slow and inaccurate,” said Centre Director Dr Anil Madhavapeddy. “These systems have undermined trust in NbS carbon credits. What is needed is a decentralised marketplace where purchasers of carbon credits can confidently and directly fund trusted nature-based projects. And that’s the gap the Centre is aiming to fill.”

̽��ֱ��Centre will support 12 PhD students and postdoctoral fellows, and investment to prototype a scalable, trusted NbS marketplace. Researchers funded from the Centre will come from the Departments of Computer Science and Technology, Zoology, and Plant Sciences, as well as from the Centre for Doctoral Training in Artificial Intelligence for the study of Environment Risk.

Professor David Coomes, Director of the ̽��ֱ�� of Cambridge Conservation Research Institute, said: “Conservation strategies are increasingly broadening to include large datasets, remote sensing technologies and computational approaches. ̽��ֱ��Centre for Carbon Credits is a ground-breaking initiative that will bring together computer scientists and conservation scientists in a new way.”

Andrew Balmford, Professor of Zoology, said: “ ̽��ֱ��recent announcement at COP26 of the new commitment to halt and reverse forest loss and land degradation by 2030 demonstrates the crucial role forests play in carbon capture and the health of our planet. ̽��ֱ��new Centre has a significant role to play in supporting crucial research to develop new, trusted mechanisms to support reforestation projects.”

Speaking on the collaborative nature of the Centre, Professor Ann Copestake, Head of the Department of Computer Science and Technology, said: “In the last few years, we’ve been expanding our emphasis on the use of computer science techniques and technologies to help address the climate emergency and the crisis in biodiversity. We are delighted to be bringing our research strengths together with the expertise in environmental science across the ̽��ֱ�� of Cambridge. We hope the work resulting from this interdisciplinary collaboration will lay the foundation for tangible solutions to some of the environmental challenges facing the world.”

A new Cambridge centre will bring together computer scientists and conservation scientists to build a trusted marketplace for carbon credits and support global reforestation efforts, the first initiative of its kind in the UK.

What's needed is a decentralised marketplace where purchasers of carbon credits can confidently and directly fund trusted nature-based projects. And that’s the gap the Centre is aiming to fill

Anil Madhavapeddy

kazuend via Unsplash

View of forest

Yes

‘Multiplying’ light could be key to ultra-powerful optical computers

sc604 — Mon, 08 Feb 2021 10:26:02 +0000

An important class of challenging computational problems, with applications in graph theory, neural networks, artificial intelligence and error-correcting codes can be solved by multiplying light signals, according to researchers from the ̽��ֱ�� of Cambridge and Skolkovo Institute of Science and Technology in Russia.

In a paper published in the journal Physical Review Letters, they propose a new type of computation that could revolutionise analogue computing by dramatically reducing the number of light signals needed while simplifying the search for the best mathematical solutions, allowing for ultra-fast optical computers.

Optical or photonic computing uses photons produced by lasers or diodes for computation, as opposed to classical computers which use electrons. Since photons are essentially without mass and can travel faster than electrons, an optical computer would be superfast, energy-efficient and able to process information simultaneously through multiple temporal or spatial optical channels.

̽��ֱ��computing element in an optical computer – an alternative to the ones and zeroes of a digital computer – is represented by the continuous phase of the light signal, and the computation is normally achieved by adding two light waves coming from two different sources and then projecting the result onto ‘0’ or ‘1’ states.

However, real life presents highly nonlinear problems, where multiple unknowns simultaneously change the values of other unknowns while interacting multiplicatively. In this case, the traditional approach to optical computing that combines light waves in a linear manner fails.

Now, Professor Natalia Berloff from Cambridge’s Department of Applied Mathematics and Theoretical Physics and PhD student Nikita Stroev from Skolkovo Institute of Science and Technology have found that optical systems can combine light by multiplying the wave functions describing the light waves instead of adding them and may represent a different type of connections between the light waves.

They illustrated this phenomenon with quasi-particles called polaritons – which are half-light and half-matter – while extending the idea to a larger class of optical systems such as light pulses in a fibre. Tiny pulses or blobs of coherent, superfast-moving polaritons can be created in space and overlap with one another in a nonlinear way, due to the matter component of polaritons.

“We found the key ingredient is how you couple the pulses with each other,” said Stroev. “If you get the coupling and light intensity right, the light multiplies, affecting the phases of the individual pulses, giving away the answer to the problem. This makes it possible to use light to solve nonlinear problems.”

̽��ֱ��multiplication of the wave functions to determine the phase of the light signal in each element of these optical systems comes from the nonlinearity that occurs naturally or is externally introduced into the system.

“What came as a surprise is that there is no need to project the continuous light phases onto ‘0’ and ‘1’ states necessary for solving problems in binary variables,” said Stroev. “Instead, the system tends to bring about these states at the end of its search for the minimum energy configuration. This is the property that comes from multiplying the light signals. On the contrary, previous optical machines require resonant excitation that fixes the phases to binary values externally.”

̽��ֱ��authors have also suggested and implemented a way to guide the system trajectories towards the solution by temporarily changing the coupling strengths of the signals.

“We should start identifying different classes of problems that can be solved directly by a dedicated physical processor,” said Berloff, who also holds a position at Skolkovo Institute of Science and Technology. “Higher-order binary optimisation problems are one such class, and optical systems can be made very efficient in solving them.”

There are still many challenges to be met before optical computing can demonstrate its superiority in solving hard problems in comparison with modern electronic computers: noise reduction, error correction, improved scalability, guiding the system to the true best solution are among them.

“Changing our framework to directly address different types of problems may bring optical computing machines closer to solving real-world problems that cannot be solved by classical computers,” said Berloff.

Reference:
Nikita Stroev and Natalia G. Berloff. ‘Discrete Polynomial Optimization with Coherent Networks of Condensates and Complex Coupling Switching.’ Physical Review Letters (2021). DOI: 10.1103/PhysRevLett.126.050504

New type of optical computing could solve highly complex problems that are out of reach for even the most powerful supercomputers.

Gleb Berloff, Hills Road Sixth Form College

Artist's impression of light pulses inside an optical computer

Yes

Women in STEM: Agnieszka Słowik

sc604 — Thu, 16 Jan 2020 07:00:00 +0000

Broadly, my research explores the reasoning capacity of neural networks. You might have seen these algorithms in action when using automatic face recognition on social media or issuing voice commands to your phone. Neural networks, also hidden behind media-friendly terms such as deep learning, are nowadays a go-to research direction if one is interested in attaining the state-of-the-art accuracy on a classification task associated with a large amount of data.

Despite their impressive practical performance, these models are limited in their ability to combine familiar ideas to arrive at new conclusions as they tend to simply memorise the data. Having learned from the examples of red squares and blue circles, a truly intelligent system surely shouldn’t be confused by a red circle. This is a core challenge in learning algorithms and I hope my research will contribute to the international efforts of the machine learning community to induce reasoning and generalisation in neural networks.

During my current internship at Mila Quebec AI Institute, I'm investigating how agents based on neural networks communicate with each other in order to solve simple games. These games draw inspiration from the studies on language evolution in humans. ̽��ֱ��communication aspect is particularly cool and exciting because by analysing the messages sent between the agents I can see how closely these algorithms mimic the reasoning process of a biological intelligent system.

I have been extremely fortunate with my supervisors (Mateja Jamnik and Sean Holden) as well as the welcoming and friendly nature of the Department of Computer Science and Technology. Cambridge provides students with a unique degree of freedom, independence and intellectual stimulation. What I particularly appreciate after my experience with competitive institutions in Poland and France is that Cambridge provides the best resources for obtaining a well-rounded education alongside the ‘hard skills’ in a student’s chosen field.

I’ve always liked the quote “the areas in which you struggle the most are the ones in which you have the most to give.” If you put a lot of effort into grasping a subject or solving a task that seems daunting to begin with, you are well-equipped to support others who struggle with the same task. I believe this also applies to challenges outside of research.

Embrace stepping out of the ‘good student’ role. ̽��ֱ��skills required in a research career, especially in science and technology, frequently won’t fully overlap with what led you to have the top grades in your previous education. Firstly, there won’t be nearly as much of the immediate positive feedback so it is crucial to enjoy the process apart from the results. Secondly, the work will never seem finished so it is more important to follow a healthy routine. Reach out to friendly experienced colleagues to find out how they cope with these challenges.

Work with a light and kind attitude to yourself and others. ̽��ֱ��trap of oscillating between imposter syndrome and ‘I’m like, a genius’ is real in research. At the end of the day you are learning, trying new things and having lots of fun, together with like-minded people.

Agnieszka Słowik is a PhD candidate in the Department of Computer Science and Technology, where she is a member of the artificial intelligence research group. Here, she tells us about neural networks and how they communicate with each other, the importance of supportive supervisors, and how to be a supportive team member.

Yes

Cambridge appoints first DeepMind Professor of Machine Learning

sc604 — Tue, 17 Sep 2019 23:09:13 +0000

Professor Lawrence joins the ̽��ֱ��’s Department of Computer Science and Technology from Amazon Cambridge, where he has been Director of Machine Learning for the past three years. He is also Professor of Machine Learning at the ̽��ֱ�� of Sheffield, where he will retain a visiting position.

Professor Lawrence’s research interests are in probabilistic models with applications in computational biology, personalised health and developing economies. At Sheffield, he led the ML@SITraN group, and helped to develop an Open Data Science Initiative an approach to data science designed to address societal needs.

“There’s so much expertise at Cambridge, in all aspects of systems and data: that’s why I’m so excited about joining,” Lawrence said. “AI and machine learning have the potential to reshape almost every aspect of our lives, but we desperately need more machine learning specialists, or else the promise of AI will not be realised.”

Professor Lawrence completed his PhD at Cambridge’s Department of Computer Science and Technology in 2000. He has previously held positions at Microsoft Research Cambridge and the ̽��ֱ�� of Manchester. In addition to his academic research, he hosts the Talking Machines podcast and is a contributor to the Guardian.

For the past five years, Professor Lawrence has been working with Data Science Africa, an organisation looking to connect machine learning researchers in Africa in order to solve problems on the ground. Professor Lawrence has an advisory role with the group, and says that many of the machine learning approaches used in Africa can have benefits in the developed world as well.

“With data and machine learning, you can have a more advanced data infrastructure in Africa than in some developed countries,” he said. “It’s rare in the UK or Europe that you’re asked to look at a machine learning problem from end to end, but you can do that in Africa, and it leads to better solutions. That’s the kind of approach I want to take to machine learning in my work at Cambridge.”

Demis Hassabis, co-founder and CEO, DeepMind, said: “I’m delighted to see Cambridge announce its first DeepMind Professor of Machine Learning. Professor Lawrence’s work in computational biology and his thoughtful advocacy for advancing technology in the developing world have been commendable. It’s an honour for DeepMind to be able to support the Department of Computer Science and Technology - from which I gained so much - in this way, and I look forward to seeing machine learning and AI flourish at Cambridge.”

“Neil will have a transformative effect on machine learning and artificial intelligence research at Cambridge,” said Professor Ann Copestake, Head of the Department of Computer Science and Technology. “He will build on our existing strengths in this area, and work with colleagues from across the ̽��ֱ�� to develop new solutions in ethical and sustainable ways.”

“It is vital we have a deep pool of talented scientists in universities and industry so the UK can continue to be a world leader in artificial intelligence,” said Minister for Digital Mark Warman. “This Government is investing millions into skills and talent training, including a number of Turing AI Fellowships in partnership with ̽��ֱ��Alan Turing Institute, and I welcome the appointment of Professor Neil Lawrence as the inaugural DeepMind Professor of Machine Learning at Cambridge. This is one of a range of moves demonstrating the enormous strength of the UK’s research base.”

In addition to the gift to support the DeepMind Professorship, the company are also supporting four Master’s students from underrepresented groups wishing to study machine learning and computer science at Cambridge. ̽��ֱ��first students supported through this programme will be starting their studies this coming term.

Following an international search, Professor Neil Lawrence has been appointed as the inaugural DeepMind Professor of Machine Learning at Cambridge, supported by a benefaction from the world-leading British AI company.

There’s so much expertise at Cambridge, in all aspects of systems and data: that’s why I’m so excited about joining

Neil Lawrence

Yes

‘ ̽��ֱ��Next Leap Forward’ – four quantum technologies hubs to lead UK’s research drive

Anonymous — Fri, 12 Jul 2019 11:57:52 +0000

̽��ֱ��National Quantum Technologies Programme, which began in 2013, has now entered its second phase of funding, part of which will be a £94 million investment by the UK government, via UK Research and Innovation’s (UKRI) Engineering and Physical Sciences Research Council (EPSRC), in four Quantum Technologies Research Hubs.

̽��ֱ�� ̽��ֱ�� of Cambridge is a partner in the Quantum Communications Hub, led by the ̽��ֱ�� of York, which is pursuing quantum communications at all distance scales, to offer a range of applications and services and the potential for integration with existing infrastructure.

Through these Hubs, the UK’s world-leading quantum technologies research base will continue to drive the development of new technologies through their networks of academic and business partnerships.

“Harnessing the full potential of emerging technologies is vital as we strive to meet our Industrial Strategy ambition to be the most innovative economy in the world,” said Science Minister Chris Skidmore. “Our world-leading universities are pioneering ways to apply quantum technologies that could have serious commercial benefits for UK businesses. That’s why I am delighted to be announcing further investment in Quantum Technology Hubs that will bring academics and innovators together and make this once-futuristic technology applicable to our everyday lives.”

“ ̽��ֱ��UK is leading the field in developing Quantum Technologies and this new investment will help us make the next leap forward in the drive to link discoveries to innovative applications. UKRI is committed to ensuring the best research and researchers are supported in this area,” said Professor Sir Mark Walport, Chief Executive of UKRI.

̽��ֱ��Quantum Communications Hub has already established the UK's first quantum network, the UKQN. They will be extending and enhancing the UKQN, adding function and capability, and introducing new Quantum Key Distribution (QKD) technologies - using quantum light analogous to that used in conventional communications, or using entanglement working towards even longer distance fibre communications.

“We will be extending the UKQN to a national scale, with links over the EPSRC National Dark Fibre Facility to London and Bristol, as well as a link to our industrial partner BT in Adastral Park in Ipswich,” said Professor Richard Penty from the Department of Engineering. “We will be using this network to trial more advanced quantum communications technologies, including quantum repeaters, quantum entanglement, continuous variable QKD and new algorithms.”

Although widely applicable, key-sharing does not provide a solution for all secure communication scenarios. ̽��ֱ��Hub will research other new quantum protocols and the incorporation of QKD into wider security solutions. Professor Adrian Kent from the Department of Applied Mathematics and Theoretical Physics is co-leading this work with other theorists in the Hub.

“We have been devising new applications of quantum communication which allow new secure cryptographic schemes, often also making use of the impossibility of faster-than-light signalling,” said Kent. “We have also been working with experimentalist colleagues in the Hub on the practical implementation of some of these schemes, for example over the UK Quantum Network.

“ ̽��ֱ��next phase of the Hub will allow us to extend our theoretical work and experimental collaborations, including work on space-based implementations via satellite links.”

̽��ֱ��Cambridge researchers will also be working on quantum communications on a chip, particularly for the networking aspects. “One of the barriers for take-up of quantum communications is that the transmitters and receivers are bespoke and made from discrete components,” said Penty. “Integrating many of the functions on the same chip will reduce the costs and speed up commercialisation.”

Given the changing landscape worldwide, it is becoming increasingly important for the UK to establish national capability, both in quantum communication technologies and their key components such as light sources and detectors. ̽��ֱ��Hub has assembled an excellent team to deliver this capability.

Adapted from a UKRI press release.

Technologies that will allow fire crews to see through smoke and dust, computers to solve previously unsolvable computational problems, construction projects to image unmapped voids like old mine workings, and cameras that will let vehicles ‘see’ around corners are just some of the developments already taking place in the UK.

Yes

Location, location, location: researchers develop model to predict retail failure

sc604 — Mon, 08 Oct 2018 23:01:44 +0000

Using information from ten different cities around the world, the researchers, led by the ̽��ֱ�� of Cambridge, have developed a model that can predict with 80% accuracy whether a new business will fail within six months. ̽��ֱ��results will be presented at the ACM Conference on Pervasive and Ubiquitous Computing (Ubicomp), taking place this week in Singapore.

While the retail sector has always been risky, the past several years have seen a transformation of high streets as more and more retailers fail. ̽��ֱ��model built by the researchers could be useful for both entrepreneurs and urban planners when determining where to locate their business or which areas to invest in.

“One of the most important questions for any new business is the amount of demand it will receive. This directly relates to how likely that business is to succeed,” said lead author Krittika D’Silva, a Gates Scholar and PhD student at Cambridge's Department of Computer Science and Technology. “What sort of metrics can we use to make those predictions?”

D’Silva and her colleagues used more than 74 million check-ins from the location technology platform Foursquare from Chicago, Helsinki, Jakarta, London, Los Angeles, New York, Paris, San Francisco, Singapore and Tokyo; and data from 181 million taxi trips from New York and Singapore.

Using this data, the researchers classified venues according to the properties of the neighbourhoods in which they were located, the visit patterns at different times of day, and whether a neighbourhood attracted visitors from other neighbourhoods.

“We wanted to better understand the predictive power that metrics about a place at a certain point in time have,” said D’Silva.

Whether a business succeeds or fails is normally based on a number of controllable and uncontrollable factors. Controllable factors might include the quality or price of the store’s product, its opening hours and its customer satisfaction. Uncontrollable factors might include unemployment rates of a city, overall economic conditions and urban policies.

“We found that even without information about any of these uncontrollable factors, we could still use venue-specific, location-related and mobility-based features in predicting the likely demise of a business,” said D’Silva.

̽��ֱ��data showed that across all ten cities, venues that are popular around the clock, rather than just at certain points of day, are more likely to succeed. Additionally, venues that are in demand outside of the typical popular hours of other venues in the neighbourhood tend to survive longer. ̽��ֱ��data also suggested that venues in diverse neighbourhoods, with multiple types of businesses, tend to survive longer.

While the ten cities had certain similarities, the researchers also had to account for their differences.

“ ̽��ֱ��metrics that were useful predictors vary from city to city, which suggests that factors affect cities in different ways,” said D’Silva. “As one example, that the speed of travel to a venue is a significant metric only in New York and Tokyo. This could relate to the speed of transit in those cities or perhaps to the rates of traffic.”

To test the predictive power of their model, the researchers first had to determine whether a particular venue had closed within the time window of their data set. They then ‘trained’ the model on a subset of venues, telling the model what the features of those venues were in the first time window and whether the venue was open or closed in a second time window. They then tested the trained model on another subset of the data to see how accurate it was.

According to the researchers, their model shows that when deciding when and where to open a business, it is important to look beyond the static features of a given neighbourhood and to consider the ways that people move to and through that neighbourhood at different times of day. They now want to consider how these features vary across different neighbourhoods in order to improve the accuracy of their model.

Reference:
Krittika D’Silva et al. ‘ ̽��ֱ��Role of Urban Mobility in Retail Business Survival.’ Paper presented to the Ubicomp 2018, Singapore, 8-12 October 2018. https://ubicomp.org/ubicomp2018/program/program.html

Researchers have used a combination of location and transport data to predict the likelihood that a given retail business will succeed or fail.

One of the most important questions for any new business is the amount of demand it will receive.

Krittika D’Silva

toastbrot81

Regent Street

Yes

Licence type:

Attribution-Noncommerical